Effects of Recent Alcohol and Drug Use On Multi-Session Health Promotion Program Participation Among HIV-Infected Patients in St. Petersburg Russia

Background:  Multisession behavioral interventions can reduce health risks among HIV-infected populations, but engagement and retention in these types of interventions remains an important issue that may be particularly relevant to active substance users.

Objective: To examine whether recent heavy alcohol use and injection drug use among HIV-infected adults in St. Petersburg, Russia affects participation in multisession behavioral interventions held at an infectious disease hospital.

Method: We analyzed baseline data from participants of the HERMITAGE study, a two-armed RCT evaluating a secondary prevention intervention for HIV-infected heavy drinkers. Participants (n=700) were randomized to either a behavioral intervention or stress management control group, each with 5 sessions (2 individual and 3 group). Multiple logistic regression analyses were conducted to assess whether past 30 day heavy alcohol use or injection drug use (IDU) was negatively associated with greater intervention completion (≥ 80% sessions completed), after adjusting for demographics, randomization group, recruitment from inpatient vs. outpatient setting and health indicators.

  Results: Participants were 59% male with a median age of 29 years. 81% reported past 30 day heavy alcohol use and 42% injection drug use. A majority (60%) completed ≥ 80% of sessions. Neither heavy alcohol use (AOR=1.14, 95% CI=0.76, 1.70) nor injection drug use (AOR=0.74, 95% CI=0.53, 1.03) were significantly associated with program participation in the fully adjusted model, but recruitment from inpatient facilities was a positively related of this outcome (AOR=1.54, 95% CI=1.11, 2.13). Due to concerns that inpatient status may be in the causal pathway, an exploratory analysis was conducted in which this factor was removed as a covariate; in this subsequent model, IDU was significantly associated with worse program participation (AOR=0.69, 95% CI=0.50, 0.96) though alcohol was not.   

Conclusions:  Recent injection drug use (IDU) and heavy alcohol use were not associated with active participation in multisession health promotion programs for HIV-infected patients when held in inpatient settings. However, IDU may still be an important potential deterrent for such participation, as its association was significant when recruitment site status (i.e., inpatient) was excluded from analysis. These insights may help guide implementation procedures for future secondary prevention interventions with substance using HIV-infected populations.